We are proposing a pharmacokinetically designed double-blind placebo-controlled study of nortriptyline (NT) in post-pubertal adolescents 12-17 years old who meed RDC and DSM III criteria for major depressive disorder (MDD) assessed by instruments similar to those utilized in studies of adults and in our double-blind placebo-controlled study of prepubertal depressed subjects. Our prior studies support the hypothesis that the pharmacology of NT during adolescence is similar to that of adults and prepubertals with respect to pharmacokinetic parameters; single dose kinetics to predict steady state plasma levels and dose; stability of plasma levels over time; the time to achieve steady state; interaction with chlorpromazine; plasma level/response relationships; and cardiovascular effects, These data permit designing the proposed study in which baseline single dose kinetics would be performed to predict the dose necessary to obtain plasma levels optimal for efficacy and safety. We have developed an adolescent age appropriate dose prediction table for use in the proposed study. The latter insures that all subjects on active drug are also at therapeutic plasma levels; and corrects for the chief criticism of the only rigorous double-blind placebo-controlled study of tricyclic antidepressant (imipramine) in the pediatric age group reported to date. Adolescents on NT in our pilot studies did not develop dose limiting cardiovascular or other side effects. 60 subjects would be studied (30 active, 30 placebo) on a 10 week protocol (single-blind place phase-2weeks; doulbe-blind placebo-controlled pahse-8 weeks). During the double-blind phase, subjects on active drug would be maintained at steady state plasma levels between 60-100 ng/ml; similar to the range in our pilot work, in the prepubertal study and to the lower part of the range in adult studies. We aimed for the lower part of the adult therapeutic range to enhance cardiovascular safety. If NT is proven effective, it would be important in the validation of the current consensus that MDD can be identified in adolescents utilizing similar criteria and assessments to those in adults. Current data suggest that pediatric MDD is a chronic illness; dissimilar to the episodic pattern reported in adults; which seriously interfers with age-appropriate family, school development. Effective pharmacotherapy may not only alleviate this morbidity but also potentially lessen the markedly increasing rate of suicide during the adolescent years. If pharmacological intervention is proven effective, comparison studies with drug and non-drug modalities could be undertaken.